On-Line Modication of Saccadic Eye Movements by Retinal Signals Valérie Gaveau, Olivier Martin, Claude Prablanc, Denis Pelisson, Christian Urquizar, Michel Desmurget
Total Page:16
File Type:pdf, Size:1020Kb
On-line modication of saccadic eye movements by retinal signals Valérie Gaveau, Olivier Martin, Claude Prablanc, Denis Pelisson, Christian Urquizar, Michel Desmurget To cite this version: Valérie Gaveau, Olivier Martin, Claude Prablanc, Denis Pelisson, Christian Urquizar, et al.. On- line modication of saccadic eye movements by retinal signals. NeuroReport, Lippincott, Williams & Wilkins, 2003, 14 (6), pp.875-878. 10.1097/01.wnr.0000069964.11849.61. hal-02197095 HAL Id: hal-02197095 https://hal.archives-ouvertes.fr/hal-02197095 Submitted on 30 Jul 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. On-line modication of saccadic eye movements by retinal signals Valérie Gaveau, Olivier Martin, Claude Prablanc, Denis Pelisson, Christian Urquizar, Michel Desmurget To cite this version: Valérie Gaveau, Olivier Martin, Claude Prablanc, Denis Pelisson, Christian Urquizar, et al.. On-line modication of saccadic eye movements by retinal signals. NeuroReport, Lippincott, Williams Wilkins, 2003, 10.1097/01.wnr.0000069964.11849.61. hal-02197095 HAL Id: hal-02197095 https://hal.archives-ouvertes.fr/hal-02197095 Submitted on 30 Jul 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. On-line modi¢cation of saccadic eye movements by retinal signals Vale¤ rie Gaveau, Olivier Martin,1 Claude Prablanc, Denis Pe¤ lisson, Christian Urquizar and Michel DesmurgetCA Espace et Action, INSERM Unite¤ 534, 16 Avenue Le¤ pine, Bron F-69676; 1UFRAPS-SPM,University J.Fourier,Grenoble F-38041,France CACorresponding Author Received 2 December 2002; accepted 7 March 2003 DOI: 10.1097/01.wnr.0 0 0 0 0 6996 4.118 49.61 A saccade is a rapid shift of the position of the eyes (o 10 0 ms). the target jump in£uenced signi¢cantly the amplitude and the peak Saccades are generally considered too quick to be in£uenced by velocity of the ongoing saccade (opposite e¡ects were found for retinal signals. To address this idea, we displaced the visual target rightward and leftward jumps).Changes in saccade kinematics oc- of a rightward horizontal saccade at eye movement onset (when curred as early as 50 ms after the target jump.These results show there is suppression of conscious perception). To prevent adaptive that retinal information is processed quickly during eye move- and learning e¡ects to occur, jump saccades were always followed ments, presumably through sub-cortical pathways. NeuroReport by a random series of 10 no-jump saccades. Results indicated that 14:000^000 c 2003 Lippincott Williams & Wilkins. Key words: Feedback; Human; Motor control; Retinal feedback; Saccade; Superior colliculus INTRODUCTION could change undetectably during the saccade (due to Saccades are rapid eye movements that shift the point of saccadic suppression, subjects were not aware of this jump). gaze from one position to another. These movements are The authors noted that the saccadic response was modu- known to be very accurate and remarkably quick (201 lated by the nature of the intra-saccadic target jump: the saccades can occur in o 70 ms [1]). Based on these amplitude of the primary saccade was slightly increased or characteristic features, most researchers agree that saccadic decreased when the target jumped forward or backward, eye movements are ballistic, i.e. independent of retinal respectively. However, this observation was questioned feedback. For example, Becker [1] noted that a retinal because of the methodological limitations related to the comparison between the current and desired eye position, recording technique (EOG) and because of the character- even if taken right at the beginning of a saccade, would be istics of the task which involved a simultaneous eye-hand too late to influence the course of the saccade. In apparent movement. A similar modulation of the saccadic response contradiction with this claim, on-line corrections in saccades was observed, by our group, in two subsequent experiments were however reported in some studies. For instance, it was using identical experimental designs (unpublished results). shown that the direction or amplitude of visually directed To specifically investigate the possibility that saccadic eye saccades could change in-flight when vertical target steps movements can be modified in response to a change of (22.51) were delivered during the reaction time of large intra-saccadic visual information, we designed an experi- horizontal saccadic eye movements (501) [2], or when ment in which: (1) eye movements were recorded with a detectable target jumps were triggered during abnormally high resolution technique; (2) intra-saccadic perturbation slow saccades in spino-cerebellar patients (4–30 times was not consciously detected by the subjects; (3) intra- slower than normal [3]). Although widely acknowledged, saccadic visual perturbation of target position occurred only these data were considered too specific to seriously once every 10 saccades to remove any possible learning challenge the canonical idea that normal saccades unfolded effect and to prevent the potential influence of a large post- uninfluenced by retinal signals. saccadic error on the characteristics of next saccade (this has A puzzling observation questioning the idea that normal been observed in adaptation paradigms; see [5] for a review, saccades could not be corrected on-line was reported by [6]); (4) no hand movement was associated with the saccadic Prablanc and Martin [4]. These authors studied automatic response task. limb corrections to a perturbed stimulus by asking human subjects to look and point to visual targets whose location s ED: Susan Koshy Op: gopal/padma WNR: lww_wnr_101 WNR lww_wnr_101 V. G AVEAU ETAL. MATERIALS AND METHODS beginning and the end of the primary and corrective Behavioral task: Fourteen subjects participated in this saccades were automatically detected using a velocity study. They were required to look at visual targets presented threshold procedure (30 deg/s). The results of this proce- in their peripheral visual field in an otherwise dark room. dure were verified off-line and corrected, if necessary. The Horizontal eye movements were recorded with an infrared main saccade-related parameters analyzed in this experi- optometric system (EyeLink, Ontario; nominal accuracy ment were the reaction time, the movement duration (MD), 0.11) at a frequency of 250 Hz. Eye velocity was detected on- the magnitude and the instant of occurrence of the eye peak line using a two-point central difference algorithm [7]. Red velocity (PV and IPV) and the amplitude of the primary light-emitting diodes (LEDs, diameter 5 mm) were located saccade (AMP). For each experimental group, a one-way on a semicircle centered on the subjects’ cyclopean eye ANOVA with repeated measures was used to test the effect (radius 110 cm). LEDs were placed in a range from À17.51 of the intra-saccadic target step (R vs J trials) on these (left) to 32.51 (right) in 2.51 intervals. Four LEDs were parameters. For the sake of simplicity, RÀ,JÀ and R + , J + designated as fixation LEDs (À17.5, À10, À5 and À2.51), denote R and J trials recorded in the GÀ and G + whereas the others were considered target LEDs. Each paradigms, respectively (signs À meaning backward and subject performed 200 trials in a single session. Each trial + meaning forward). unfolded as follows: (1) the subjects looked at one of the fixation LEDs (illumination period, 2 s 7 300 ms); (2) the fixation LED was turned off and one of the target LEDs was simultaneously turned on (illumination period, RESULTS 2s7 300 ms) signaling the subject to perform a rightward Statistical results and mean values for the main experi- saccade; (3) for the leftward return saccade, the target LED mental parameters are reported in Table 1. was turned off while one of the fixation LEDs was turned When questioned at the end of the experiment, only two on. of the subjects (S1 G + and S5 GÀ) were able to report the For convenience, trials were segmented into 20 blocks of occurrence of the intra-saccadic target jump. As shown 10 trials. For each block three types of trials were presented: below, the behavior of the two subjects who detected the jump (J), reference (R), or standard (S). Trial 10 was always a jump was qualitatively and quantitatively coherent with the J trial. It started with a À101 fixation point followed by an behavior of the subjects who had not detected the jump. initial target at 201. During the saccadic displacement, the At a quantitative level, no significant difference was target location was modified once eye velocity exceeded observed for RT, MD and IPV between R + and J + and 30 deg/s. In one group of subjects (GÀ; n ¼ 7) the jump was between RÀ and JÀ. In contrast, significant differences were backwards (20 -12.51) while in a second group (G + ; n ¼ 7) observed for AMP and PV (Table 1). In group G + (forward it was forward (20-27.51). Either trial 7 or trial 8 or trial 9 jump), an increase in saccadic amplitude (1.21 on average) was an R trial (random selection).